作者: Jürgen Rurainsky
DOI: 10.1007/978-3-642-02611-9_66
关键词:
摘要: Investigation of the motion performed by a person's face while speaking is target this paper. Methods and results studied facial motions are presented rigid non-rigid analyzed. In order to extract only deformation independent from head pose, we use new simple approach for separating called Weight Compensated Motion Estimation (WCME). This weights data points according their influence desired model. A synthetic test as well real used demonstrate performance approach. We also present in field analysis basis shapes description form. These can be recognition purposes adding temporal changes overall process or natural deformations other than at given database.